Towards Resilience Metrics for Future Cloud Applications

Novak, Marko; Shirazi, Syed Noorulhassan; Hudic, Aleksandar; Hecht, Thomas; Tauber, Markus; Hutchison, David; Maksuti, Silia; Bicaku, Ani

doi:10.5220/0005910502950301

Cited by 5 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Whilst universal resilience metrics suitable for arbitrary systems have not yet been developed, various specialized metrics are commonly used in domain-and layer-specific scenarios [106]. Depending on the type of systems, resilience may be achieved at different layers against different types of potential issues.…”

Section: Layer-specific Resiliencementioning

confidence: 99%

“…System resilience is commonly defined in terms of a system's ability to provide an ''acceptable level of operation'' under certain conditions [105]. Unfortunately, strict resilience metrics are not universally defined [106], [132], [142]. Once the operation level degrades below a pre-set threshold, various restoration procedures must be carried out until the desired level of operation is achieved.…”

Section: A Formal Resilience Orchestrationmentioning

confidence: 99%

See 1 more Smart Citation

Architectural Resilience in Cloud, Fog and Edge Systems: A Survey

Prokhorenko

Babar

2020

IEEE Access

View full text Add to dashboard Cite

An increasing number of large-scale distributed systems are being built by incorporating Cloud, Fog, and Edge computing. There is an important need of understanding how to ensure the resilience of systems built using Cloud, Fog, and Edge computing. This survey reports the state-of-the-art of architectural approaches that have been reported for ensuring the resilience of Cloud-, Fog-and Edge-based systems. This work reports a flexible taxonomy for reviewing architectural resilience approaches for distributed systems. In addition, this work also presents a capability-based cyber-foraging framework intended to improve the overall system resilience in the context of a physical node's capabilities. This survey also highlights the trust-related issues and solutions in the context of system resilience and reliability. This survey will help improve the understanding of the current state of system resilience solutions and raise awareness about the issues related to physical capabilities and trust management in the context of distributed systems resilience.

show abstract

Section: Layer-specific Resiliencementioning

confidence: 99%

Section: A Formal Resilience Orchestrationmentioning

confidence: 99%

Architectural Resilience in Cloud, Fog and Edge Systems: A Survey

Prokhorenko

Babar

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The potential resilience issues are relevant to security issues, e.g., sniffing, session hijacking and flooding attacks. The reliability metrics/attributes are: mean time to failure and mean time to repair, but other attributes also affect service directly or indirectly [34].…”

mentioning

confidence: 99%

Network Slicing Security Controls and Assurance for Verticals

et al. 2022

View full text Add to dashboard Cite

This paper focuses on the security challenges of network slice implementation in 5G networks. We propose that network slice controllers support security by enabling security controls at different network layers. The slice controller orchestrates multilevel domains with resources at a very high level but needs to understand how to define the resources at lower levels. In this context, the main outstanding security challenge is the compromise of several resources in the presence of an attack due to weak resource isolation at different levels. We analysed the current standards and trends directed to mitigate the vulnerabilities mentioned above, and we propose security controls and classify them by efficiency and applicability (easiness to develop). Security controls are a common way to secure networks, but they enforce security policies only in respective areas. Therefore, the security domains allow for structuring the orchestration principles by considering the necessary security controls to be applied. This approach is common for both vendor-neutral and vendor-dependent security solutions. In our classification, we considered the controls in the following fields: (i) fair resource allocation with dynamic security assurance, (ii) isolation in a multilayer architecture and (iii) response to DDoS attacks without service and security degradation.

show abstract